INDIANA UNIVERSITY - Weebly
INDIANA UNIVERSITY
The David Skomp Distinguished Lectures in Anthropology
1998–99
EVOLUTIONARY MEDICINE AND BREASTFEEDING: IMPLICATIONS FOR RESEARCH AND PEDIATRIC ADVICE
Katherine A. Dettwyler Text of a lecture
Texas A&M University delivered November 19, 1998
INDIANA UNIVERSITY DEPARTMENT OF ANTHROPOLOGY
The Skomp Distinguished Lecture Series in Anthropology is made possible by an endowment provided to the Indiana University Anthropology Department in 1983 by David Skomp (AB 1962; MS 1965). Mr. Skomp studied under the direction of Dr. Georg Neumann. He left the remainder of his estate for the use of the Department. The Skomp Endowment currently provides support for anthropology students in the form of first-year fellowships, summer field research grants, dissertation-year fellowships, and distinguished lectures such as these.
EVOLUTIONARY MEDICINE AND BREASTFEEDING: IMPLICATIONS FOR RESEARCH AND PEDIATRIC ADVICE
Katherine A. Dettwyler
ABSTRACT—Evolutionary medicine seeks to understand modern human health and disease in light of the conditions under which humans evolved, and recognizes that modern lifestyles and cultural beliefs may conflict with our underlying mammalian, primate, and hominid evolutionary legacies. Breastfeeding and weaning behaviors are prime examples of areas where modern cultural beliefs and practices, especially in the United States, conflict with underlying adaptations. In the United States, many children are not breastfed at all, or are breastfed for only a few weeks or months; around the world the incidence and duration of breastfeeding are declining, and one set of researchers have called for international health organizations to establish recommended upper limits on duration of breastfeeding. In the literature, nursing beyond six months to one year is often described as “prolonged” or “extended” breastfeeding. In contrast, recent research examining nonhuman primate life history variables that are closely correlated with duration of breastfeeding suggests that the natural duration of breastfeeding for modern humans is somewhere between 2.5 years and seven years, with many predictors clustering at the five- to six-year end of the range. For example, among non-human primates, isochrony of weaning with the eruption of the first permanent molar predicts 5.5–6.5 years of breastfeeding for humans; the relationship of weaning to reproductive maturity predicts almost five years of breastfeeding for a primate species with first breeding at 16 years; the relationship of weaning to gestation in large-bodied hominoids predicts 4.5 years of breastfeeding for humans. The duration of breastfeeding directly affects human health over the entire life span, with children breastfed the longest having the lowest risk for many diseases and conditions, and mothers who breastfed the longest having the lowest risk of breast cancer. Duration of breastfeeding is the only life span stage amenable to direct and substantial cultural manipulations, and the only one for which routine shortening or complete elimination has become accepted as the norm in some cultures. It is the only life history variable for which maternal behaviors that match evolutionary expectations have been culturally defined as abnormal and suspect. Research suggesting that prolonged breastfeeding results in malnutrition and poor growth has failed to recognize that children who have died due to premature weaning cannot contribute their anthropometric measurements to the “already weaned” sample. The recognition that 2.5 to seven years of breastfeeding is what modern human children have evolved to expect opens many new areas for research, has important implications for pediatric advice and legal issues, and provides a different perspective on studies of “prolonged breastfeeding.” This information also raises questions about the normality and healthiness of common behaviors of bottle-fed children such as thumb-sucking and attachment to inanimate objects.
Key words—evolutionary medicine, breastfeeding, weaning, life history, infant feeding, child health
INTRODUCTION
Evolutionary medicine seeks to understand modern human health and disease in light of the conditions under which humans evolved, and recognizes that modern lifestyles and cultural beliefs may conflict with our underlying mammalian, primate, and hominid evolutionary legacies. Breastfeeding is both a biological process and a heavily culturized activity. As a biological process, breastfeeding is firmly grounded in our mammalian ancestry and absolutely critical to the survival of the species. As a heavily culturized activity, breastfeeding is modified by a variety of beliefs about infant/child health and nutrition, the nature of human infancy, and the proper relationships between mother and child, mother and father, and mother and society. In addition, breastfeeding practices are affected by such disparate influences as religious beliefs, the structure of the mother’s everyday work activities, the health of the child, and seemingly unrelated ideas about personal independence and autonomy. Thus, the timing of the end of breastfeeding—weaning—has physiological underpinnings, the result of millions of years of evolutionary adaptation, but is also subject to cultural modifications. Breastfeeding and weaning behaviors are prime examples of areas where modern cultural beliefs and practices conflict with underlying adaptations. In the United States, many children are not breast-fed at all, or are breastfed for only a few weeks or months (Ross Labs, 1993). Around the world the incidence and duration of breastfeeding are declining, and one set of researchers (Nubé and Asenso-Okyere, 1996) have called for international health organizations to establish recommended upper limits on duration of breastfeeding, to protect children from the supposed health hazards of prolonged breastfeeding. More and more frequently, a strategy in U.S. child-custody cases is for the husband to accuse the mother of sexually abusing her child by breastfeeding (Elizabeth Baldwin, pers. comm., and the author’s consulting phone logs). At the same time, recent research examining nonhuman primate life history variables that are closely correlated with duration of breastfeeding suggests that the natural duration of breastfeeding for modern humans is somewhere between 2.5 years and seven years, with many predictors clustering at the five- to six-year end of the range (Dettwyler, 1995a). To understand the potential for a mismatch between evolutionary expectations and modern cultural realities, we must first review the evidence for a natural or physiological duration of breastfeeding in modern humans. Thus, the primary purpose of this paper is to summarize the research that examines age at weaning within an evolutionary framework, and to discuss the implications of this evolutionary perspective for pediatric advice and for future research on the consequences of premature weaning for women’s and children’s health. The conclusions of Nubé and Asenso-Okyere (1996), that prolonged breastfeeding can be harmful to children’s growth and nutritional status, are shown to be inaccurate, and due to their failure to account for differential mortality in their analyses of their data.
PRIMATE LIFE HISTORY VARIABLES
The answer to the question of how long human children would breastfeed (or be breastfed) if they followed a natural or physiological pattern, not influenced by cultural beliefs, lies in an examination of the nonhuman primate data on life history variables. Life history variables refer to the length and characteristics of various stages of the primate life span. Traditionally, life history variables begin with the prenatal period, defined as the average length of gestation. The next stage is infancy, traditionally defined as lasting from birth until the eruption of the first permanent teeth. The juvenile stage lasts from the eruption of the first permanent tooth until the eruption of the last permanent tooth. The adult period begins with the eruption of the last permanent tooth, and continues to the end of mean longevity for the species. The reproductive period in females can be defined in various ways; often data are available on average age of reproductive maturity for females as measured by menarche (first menstrual cycle or first estrous), and average age at first breeding or birth. The postreproductive period in females lasts from menopause until the end of mean longevity. Life history variables also include various size and growth rate variables, such as adult female weight, adult male weight, birth weight, number of offspring born at one time, weaning age [duration of breastfeeding], length of the estrous cycle, interbirth interval, brain weight at birth, brain weight in adults, and rates of growth as measured by either multiples of birth weight or percent completion toward adult weight. Research has shown that many life history variables are strongly correlated with each other, and with adult body size.
Members of the Order Primates have longer stages of the life span, and longer life spans, than members of other Orders of placental mammals. Within the Order, several trends can be identified, most of which are related to body size. Comparing the durations of life stages in lemurs (representing prosimians), macaques (Old World Monkeys), gibbons (Lesser Apes), chimpanzees (Great Apes), and humans, reveals a clear pattern. As one moves from the simpler, smaller, mostly nocturnal and often solitary prosimians up through the more complex, larger, diurnal, more intelligent and usually more social higher primates, every segment of the life span is elongated. Gestation length increases from 18 weeks in the lemur to 24 in the macaque, 30 in the gibbon, 33 in the chimpanzee, and 38 in modern humans. Duration of other segments of the life span likewise increases, from eruption of the first permanent teeth to eruption of the last permanent teeth, to duration of the reproductive period, and finally elongation of the postreproductive period in humans, resulting in the longest average life spans for humans. In other words, every segment of the life span is elongated in primates compared with other placental mammals, and every segment of the life span not subject to cultural manipulation is elongated in humans compared with the other primates.
Harvey and Clutton-Brock found that “larger primates wean their offspring later relative to their body size than is the case for other mammalian orders” (1985:577). Within the order, primate subfamilies that have relatively large neonates (high birth weight to adult weight ratios) have relatively long gestations, late age at sexual maturity, long life spans, and large neonatal and adult brain sizes, when compared with subfamilies that have relatively small neonates (Harvey and Clutton-Brock, 1985). Additionally, primate subfamilies with relatively large offspring wean them at a later age than subfamilies with relatively small offspring. Thus, we would expect humans, as large primates with relatively large neonates, to have among the latest weaning ages of the order.
PREDICTIONS OF THE NATURAL DURATION OF BREASTFEEDING IN HUMANS
In this paper, the term weaning refers to the complete cessation of breastfeeding, i.e., the time when the child is no longer allowed, or no longer wants, to nurse from its mother’s breasts. The use of the term weaning to denote complete cessation of breastfeeding recognizes that non-milk foods will be added to the diet of the child long before weaning takes place.
In the sections that follow, I summarize the comparative primate data concerning the relationship between age at weaning and: (1) adult female body weight, (2) attainment of one-third adult weight, (3) growth rates, based on multiples of birth weight, (4) length of gestation, (5) age at eruption of the first permanent molars, and (6) age at first breeding. Where appropriate, determinations based on data from modern human populations of various adult body sizes are presented.
Weaning according to adult female body weight
In 1985, Harvey and Clutton-Brock published their compilation of comparative life history data for 135 primate species. They examined the relationships among such variables as adult body weight (by sex), gestation length, birth weight, number in litter, weaning age, age at sexual maturity, interbirth intervals, neonatal and adult brain weights, and others. Harvey and Clutton-Brock found that many life history variables were closely tied to average adult body size within each subfamily. From their data, they derived regression equations for the prediction of the various life history variables as a function of adult female body weight. Their equation for calculation of weaning age is:
weaning age in days = 2.71 X adult female body weight in grams.56
One problem with using this equation on human population is how to define “adult weight,” since human populations occupy many different ecological niches, and exhibit greater variation in average body weight than nonhuman primate species that live within a restricted environmental range. Rather than pick one population as representative of all humans, several different human populations will be examined to include the range of variation in adult female body weight in modern humans. For the small-bodied !Kung (data from Truswell and Hansen, 1976), with an average adult female body weight of 40.10 kg, the regression equation predicts an age at weaning of 2.80 years (1,022 days). For the medium-bodied rural Malians (ethnically Bambara, data from Dettwyler, 1992), with an average adult female body weight of 53.60 kg, the regression equation predicts an age at weaning of 3.30 years (1,205 days). For the large-bodied United States population (NCHS data, Hamill et al., 1979), with an average adult female body weight of 55.35 kg, the regression equation predicts an age at weaning of 3.36 years (1,228 days). For the extra-large-bodied Inuit (data from Jamison, 1978), with an average adult female body weight of 64.50 kg, the regression equation predicts an age at weaning of 3.66 years (1,338 days).
Thus, using Harvey and Clutton-Brock’s equation, a natural age at weaning in modern humans would fall between 2.8 and 3.7 years, depending on average adult female body weight (see Table 1).
Weaning according to attainment of one-third adult weight
Additional data are provided by looking at weaning and progress toward attaining adult weight. The exact nature of the link between age at weaning and rate of growth is not known, but seems strong across a variety of mammalian species. According to Charnov and Berrigan, “On average, primates are like other mammals in weaning each offspring when they reach about one-third their adult weight.” (Charnov and Berrigan, 1993:192, citing Charnov, 1991 and 1993). What does this mean for humans?
The natural ages at weaning for Inuit (extra-large), U.S. (large), and Bambara (medium) children predicted by growth patterns and adult body size are remarkably consistent (see Table 1). Across these three populations with very different adult body sizes, children reach one-third their adult weight at approximately seven years of age for males, and six years of age for females. In the smaller-bodied !Kung populations, one-third adult body weight is reached earlier, at five to six years for males, and four to five years for females. Following Charnov and Berrigan’s (1993) suggestion that primates generally wean their offspring when they reach one-third adult weight, then four to seven years of breastfeeding is the appropriate range for Homo sapiens, with boys generally being weaned later than girls.
Weaning according to specific multiplication of birth weight
Lawrence’s Breastfeeding: A Guide for the Medical Profession, provides a handy rule of thumb for predicting a natural age of weaning from human growth rates: “When weaning time is correlated with birth weight in placental mammals, a ratio of 3:1 is noted, i.e., weaning takes place when birth weight has tripled.“ (1994:311, emphasis added).
Children in Western industrial populations usually triple their birth weight by one year of age; thus, Lawrence implies that weaning at one year would be normal and natural for modern humans. Many pediatricians use this guideline to recommend weaning by one year of age. However, Lawrence’s “rule of thumb” does not hold up to scrutiny. Research by Lee, Majluf, and Gordon (1991) has reexamined the evidence for linkages between age at weaning and the attainment of “a critical or threshold body weight attained by offspring among large-bodied mammals: the anthropoid primates, ungulates, and pinnipeds” (1991:99). Among these large-bodied mammals, Lee and colleagues found, contra Lawrence, that weaning occurred when offspring had quadrupled their birth weight, whatever the length of time it took to achieve this milestone of growth. Specifically, they reported: “Weaning weight appeared to be a relatively constant proportion of neonatal weight such that when a weight of around four times birth weight is reached infants are weaned, irrespective of the time taken to achieve weaning.” (Lee, Majluf and Gordon, 1991:104).
Thus, the relationship between growth and weaning in large mammals suggests that quadrupling of birth weight, rather than tripling, may be a natural trigger of weaning. When do human children quadruple their birth weight? Data based on U.S. populations (NCHS, Hamill et al., 1979) and data collected by the author near Bamako, Mali, in West Africa (Dettwyler, 1985) provide two specific examples.
According to the NCHS standards, the 50th percentile for birth weight for males is 3.27 kg (7 lbs. 3 oz.). A quadrupling of that birth weight, to 13.08 kg (28 lbs. 12 oz.), occurs at around 27 months of age (50th percentile). For females, the figures are 3.23 kg at birth, with a quadrupling of birth weight to 12.92 kg at around 30 months. Thus, for Western industrialized countries, a quadrupling of birth weight usually is achieved sometime between two and three years of age.
The data from Mali, West Africa, are based on a much smaller sample and come from a periurban population. Most of these children have little or no access to modern medical care, have no access to immunizations, and suffer from many diseases, particularly measles, malaria, and gastrointestinal and upper respiratory infections. As in many other Third World populations, growth in these children is generally good for the first six months of life, but falls away sharply from the NCHS standards during the latter part of the first year, and in the second and third years of life. According to data collected by the author in 1981–83, average birth weight for males was 3.12 kg. A quadrupling of that average, to 12.48 kg, had not been reached by three years of age (Dettwyler, 1985:255), when the average weight was only 11.58 kg. For females, the figures were 2.78 kg at birth and 11.12 kg at quadrupling, an average value reached between 2.5 and 3.0 years.
In their study, Lee, Majluf, and Gordon report that in many primate species, quadrupling of birth weight correlates with return of fertility in the mother, but that this “. . . generally does not correspond to ‘weaning’ defined as the cessation of suckling, since suckling at low levels in many species continues through pregnancy until subsequent parturition” (Lee, Majluf, and Gordon, 1991:101). Thus, any estimated age at weaning in primates, if it means complete cessation of breastfeeding, should include the time it takes for quadrupling of birth weight, plus some additional months of “suckling at low levels.” The data provided by Lee, Majluf, and Gordon (1991) allow us to establish a natural age of weaning (complete cessation of breastfeeding) in humans as some months after quadrupling of birth weight, which would be close to three years of age for well-nourished, healthy populations, and between three and four years of age for marginally nourished populations dealing with multiple environmental stresses (see Table 1).
Weaning according to gestation length
It is often reported in the breastfeeding literature that a natural weaning age would be one that corresponded to the length of gestation. For example, Lawrence writes: “As a general rule, the smaller the animal, the shorter the time required for both gestation and maturation of the young. The weaning process is a gradual one, terminating after a time approximately equal to the period of gestation. The elephant’s gestational period is 20 to 21 months, and the young are totally weaned at about two years of age.” (Lawrence, 1994:311, emphasis added).
The clear implication for humans is that weaning should be expected to take place after only nine months of breastfeeding, and this relationship is often cited by pediatricians to justify and legitimize their advice to mothers to stop breastfeeding their children at nine months. Data from nonhuman primates suggest that the relationship between gestation length and age at weaning is not one of equivalence, and that the relationship between these two life history variables is heavily dependent on adult body size. Harvey and Clutton-Brock provide data on life history variables for 135 primate species (1985:562–566, their Table 1); 36 entries include information on both length of gestation and weaning age. The data for gestation length and weaning age in these 36 species are presented in Figure 1 (data extracted from Harvey and Clutton-Brock, 1985), arranged according to increasing adult female body weight.
As can be seen from Figure 1, among the smaller-bodied primates, the weaning age to length of gestation ratio is less than 1.00 (the duration of breastfeeding is shorter than the length of gestation). For all of the larger-bodied primates (monkeys and apes), the weaning age to length of gestation ratio is greater than 1.00 (the duration of breastfeeding is longer than the length of gestation). Among the Great Apes, the closest living relatives of humans, the ratios are 4.21 for Pongo pygmaeus (orangutan), 6.18 for Gorilla (gorilla), and 6.40 for Pan troglodytes (common chimpanzee). Thus, among large-bodied primate species, the average duration of breastfeeding far exceeds the average length of gestation. For humankind’s closest relatives, chimpanzees and gorillas, the duration of breastfeeding is more than six times the length of gestation. Humans are among the largest of the primates, and share more than 98% of their genetic material with chimpanzees and gorillas. Interpolating from these comparisons, an estimated natural age at weaning for humans would be a minimum of six times gestation length, or 4.5 years.
Weaning according to timing of eruption of the first permanent molars
Research by anthropologist Holly Smith has examined the relationship between timing of dental eruption and age at weaning in 21 different primate species (Smith, 1989, 1991a, 1991b, and 1992). Smith finds that age at weaning is not only highly correlated with age at eruption of the first permanent molar, but that these events occur simultaneously. In other words, among many primates, offspring are weaned when their first permanent molars are erupting. In chimpanzees, however, because they have relatively early eruption of the first permanent molars, first molar eruption often occurs several years before the end of breastfeeding. In modern humans, the first permanent molars erupt around 5.5 to 6.5 years of age.
Weaning according to age at first breeding for females
Harvey and Clutton-Brock’s (1985) data on nonhuman primates show a close correlation between age at weaning and age of reproductive maturity, measured either as “sexual maturity” meaning first ovulation or estrous (menarche in humans) or as “average age at first breeding for females.” Both are related to body size, with larger-bodied species having longer durations of breastfeeding and later ages at reproductive maturity. For example, gorillas breastfeed for about four to five years, and females give birth to their first offspring at about 10 years of age (Graham, 1981). Chimpanzees breastfeed for four and a half to seven years (Van Lawick-Goodall, 1973; four years is used by Harvey and Clutton-Brock), and females give birth to their first offspring at about 11 to 12 years (some sources cite 14 to 15 years). From Harvey and Clutton-Brock (1985), 28 species provide date for both age at weaning and age at first breeding for females. The correlation between these two variables is quite high, R=.89. The regression equation for predicting age at weaning from age at first breeding is:
weaning age in days = 1.4493 X age at first breeding in months1.3450
Again, we are faced with the difficulty of deciding when would be a natural or normal “age at first breeding” for modern humans. It is well documented that age at menarche (first menstrual period) becomes progressively younger as health and nutritional status during childhood improve (Eveleth and Tanner, 1990; Wyshak and Frisch, 1982). Women in populations living under adverse environmental conditions may not experience menarche until their late teens (Wood et al., 1985), while the average age of menarche for Western industrial countries seems to have stabilized at an average of about 12 to 13 years (Treloar and Martin, 1990). At the same time, age at marriage and age at first birth are highly influenced by cultural norms, which are themselves affected by political and economic conditions. If we use 20 years as the average age at first breeding for modern humans, the regression equation predicts an average duration of breastfeeding of 6.31 years. If we assume that the average age at first breeding is 16 years of age, the regression equation predicts an average duration of breastfeeding of 4.68 years. If we assume that the average age at first breeding is 12 years of age, the regression equation predicts an average duration of breastfeeding of 3.18 years. Figure 2 shows the relationship between these two variables for 28 non-human primates, and for the three hypothetical human datum points.
Given what we know about environmental conditions and levels of health and nutrition throughout human evolution, it is reasonable to assume that most females experienced menarche and first breeding in their late teens to early twenties for most of human evolutionary history. It was only with the advent of cultural developments such as the control of fire (700,000 to one million years ago, at the earliest), the domestication of plants and animals (beginning 10,000–12,000 years ago), and the beginnings of medical knowledge for prevention and treatment of diseases (an unknown date), that the average age of menarche fell much below 20 years. It is only in the last 50 years, in populations with excellent nutritional status and modern health care, that average age of menarche has fallen below 13 years. Thus, while it appears that, under conditions of improving nutritional status and health during childhood, age at weaning could undergo a negative secular trend similar to that observed for menarche, the lowest predicted duration of breastfeeding still would be more than three years for populations with potential first breeding at age 12 years.
To look at it from another perspective, primates that wean their offspring at one year of age have their first breeding at four to five years of age. Primates that wean at two years of age have their first breeding at four to nine years of age. Like other methods of predicting a natural age of weaning for modern humans, age at reproductive maturity predicts between three and six years of age.
Summary of the life history comparisons
An examination of the relationships between age at weaning and various life history variables among the nonhuman primates has revealed that, if humans weaned their offspring according to the primate pattern, without regard to cultural beliefs and customs, most children would be weaned somewhere between 2.5 and 7.0 years of age (Table 1 and Figure 3).
In humans, several physiological milestones occur around six to seven years of age, all of which can be linked logically to having breast milk as a component of the diet throughout this period. First, achievement of adult immune competence occurs at approximately six years of age. Until this time, the child’s active immune response (both serum and secretory) can be enhanced by the lymphokines in maternal milk (Hahn-Zoric et al., 1990; Pabst and Spady, 1990). Children need these lymphokines, even in small amounts, to augment and prime their own immune responses to stress until they ‘achieve adult levels of immune competence (IgA, IgG, IgM) around the age of six years (D. Fredrickson, pers. comm.). Second, recent research suggests that brain growth in weight is complete at an average of six to seven years of age (Cabana et al., 1993; Jolicouer et al., 1988). Several long-chain polyunsaturated fatty acids (LC-PUFAs), including docosahexanoic acid and arachidonic acid, are critical constituents of brain growth and retinal development (de Andraca and Uauy, 1995). These LC-PUFAs are not found in artificial infant formulas, and are found in relatively few other foods, such as tuna and other cold-water ocean fish. It makes sense that breast milk would be a part of the diet as long as the brain was still growing, especially under evolutionary conditions where the non-breast milk components of the diet were low in these nutrients and where populations were subject to periodic dietary deficiencies. Third, developmental psychologists have long noted that a qualitative change in intellectual processing takes place at around seven years of age. Piaget referred to this as a change from primarily preoperational thought to primarily concrete operational thought (Ginsburg and Opper, 1979). Fourth, Smith suggests that the eruption of the first permanent molar “should enhance a juvenile’s ability to process food” (Smith 1992:138). These factors support the idea that the first six or seven years of a human child’s life represent a time when the child is still dependent on maternal care, including maternal breast milk, for optimal development.
DISCUSSION
Cultural influences on primate life history variables
Returning to primate life history variables, and segments of the life span, some variables are clearly amenable to direct cultural manipulations, while others can only be affected indirectly, and within narrow limits, or cannot be affected at all (Table 2). Age at weaning is the life history variable most easily and most profoundly modified by cultural beliefs. While other stages of the life cycle can be shortened or lengthened, the aim is usually to lengthen them; shortening them may be viewed as morally repugnant. Age at weaning is also the only variable to which strong legal, moral, and psychological implications have been attached in Western industrialized societies.
Length of gestation is under genetic control, and varies little among different populations of the same species; it seems impervious to environmental influences and is not affected by nutritional status of the mother, for example, nor by altitude, temperature, rainfall or levels of physical activity. It is only slightly modifiable by cultural interventions, in two ways. First, induction of labor using modern pharmaceuticals can shorten the length of gestation, whether for the health of the mother or baby or, increasingly often, for the doctor’s or parents’ convenience. Second, the development of modern medical technology has allowed the survival of premature infants, thus shortening the length of gestation that can result in a viable infant. Neither of these modern technologies is typically viewed as a preferable or desirable condition, however, although premature induction of labor has become widely accepted in the United States. There are no cultures where a relatively short duration of gestation is advocated as desirable. No one suggests that a mother who carries her pregnancy to 38 or 40 weeks (or even 42 weeks) is acting in any way detrimental to her offspring, nor is she accused of fostering dependence in her child, or of having ulterior, psychologically unhealthy motives for wanting her pregnancy to continue for its normal duration. No one would suggest that a pregnancy lasting only 20 weeks would be desirable or better for either mother or child.
Eruption of the first and last permanent teeth, the life history variables that mark the end of the infant and juvenile periods, respectively, are also fairly impervious to cultural manipulation (Garn and Bailey, 1978). Whatever the nutritional status or health of the child, teeth erupt in a regular pattern of order and timing, and cannot be forced to erupt earlier or later simply because cultural beliefs suggest that early or late eruption would be more desirable. In some cultures, the permanent tooth buds are surgically removed from the gums before eruption, but there are no ethnographic cases known to me where children or their parents are rewarded or praised for early erupting teeth or punished or criticized for late erupting teeth.
The beginning of the reproductive period in females is subject to indirect cultural influence, within limits, through the effects of nutrition and health on age at menarche. This is known as the negative secular trend for age at menarche. Modern human populations that experience good childhood nutrition and relative freedom from disease have an earlier age at menarche than populations that suffer from chronic undernutrition and many diseases. For example, the average age at menarche in the U.S. has steadily fallen over the past century to its current average of 12.6 years (Treloar and Martin, 1990), while the average age at menarche among some Papua, New Guinea highland populations is relatively late, at 18.4 years (Johnson et al., 1987; Wood et al., 1985). The indirect influence of cultural beliefs and practices on age at menarche work entirely through environmental variables. There are no cultures where cultural beliefs and practices can directly affect age at menarche, nor are there any cultures where an early age at menarche is thought to be advantageous or desirable. There are no cultures where mothers are encouraged to have their daughters become reproductively mature at an early age, as a sign of their independence. On the contrary, public sentiment in the United States often laments the negative secular trend toward an ever lower age of reproductive maturity, and particularly laments the seemingly inevitable outcome of widespread early menarche, which is unwed teenage pregnancy. An American mother who encouraged her daughter to have her first child at the earliest age physiologically possible would be considered a very poor mother. Additionally, in many states in the U.S., the age at which a young woman is considered old enough to grant consent for sexual relations comes several years after the age at which she becomes reproductively mature, suggesting that delaying “breeding” until after the natural age of reproductive maturity is the desirable modification of the natural physiological process.
The end of the reproductive period in females can be manipulated by cultural beliefs within limits. Due directly to cultural beliefs, many women in Western industrialized countries artificially finish their reproductive careers long before declining fecundity and menopause makes further reproduction biologically impossible. Hysterectomy, tubal ligation, endometrial ablation, and the use of birth control pills and barrier devices can result in an early end to a woman’s reproductive career. Newly developed technologies may also allow for the possibility that postmenopausal women can continue to reproduce, and estrogen replacement therapy means that women continue to menstruate beyond the time when they are fecund. Nevertheless, the actual biological process of menopause is not easily affected by cultural beliefs alone. There are no cultures where women must bow to community pressure to either go through menopause early, or continue reproducing beyond the time when their body’s hormonal system has shut down.
Average longevity can be manipulated indirectly through better diet and medical care, especially sophisticated technological measures taken to prolong life. Again, there are no cultures where people are routinely encouraged to end their lives early, nor can people age prematurely or live longer simply because it is considered the moral or proper thing to do.
Age at weaning is the only life history variable that is subject to direct and substantial cultural intervention, and the only life span segment for which shortening the duration or eliminating it entirely has come to be viewed as normal and desirable. Almost half of all children born in the United States in the 1990s are effectively weaned at birth, since they are never put to their mother’s breast. Of the 50–60% who begin life breastfeeding, the vast majority will no longer be breastfeeding by six months postpartum. It is unknown how many children in the United States nurse until the physiological or natural age range suggested by the comparative primate data, 2.5 years to 7.0 years, because parents and children often feel compelled to hide this behavior from relatives, friends, neighbors, teachers, and researchers. Nevertheless, most children in the United States are weaned long before 2.5 years, the minimum age of natural weaning. This pattern of premature or early weaning can be traced to direct cultural influences, including the recommendations found in baby/parenting books and magazines, advice offered by pediatricians, and pressure from health care professionals, family members, coworkers, friends, and even complete strangers to end the breastfeeding relationship. In 1997, when the American Academy of Pediatrics released new guidelines recommending that all infants be breastfed for a minimum of 12 months, there was public outrage, widely reported in the media, that modern mothers should be expected to breastfeed for so long, as though 12 months of breastfeeding were way beyond the capability of women in the U.S.
Age at weaning is the only life history variable for which the normal or natural duration has been defined by many health care professionals and the public alike as pathological, abnormal, perverted, or dysfunctional. It is the only life history variable for which following the physiological pattern has been assigned negative moral and psychological connotations. It is the only life history variable for which following the physiological pattern may result in the mother being accused of harming her child’s physical and emotional health, meeting her own needs for sexual satisfaction, infantilizing her child, or of being too ‘psychologically enmeshed’ in her child (Love, 1991). Age at weaning is the only life history variable used in child custody cases or in charges of child sexual abuse to characterize normal human behavior as damaging to the child (see Dettwyler, 1995b for a more thorough discussion of these issues).
Health implications
The value of an evolutionary perspective on what would be a normal or natural duration of breastfeeding in modern humans does not lie in its power to suggest or mandate a particular duration of breastfeeding for all mothers. The evidence does suggest that 2.5 years is the minimum duration of time that the child has evolved to expect breastfeeding to continue, and that four, five, six years, or longer is within the normal range for humans. The evidence does not tell us when would be the ideal age of weaning for any particular mother/child pair. It is clear that advances in the development of nutritionally adequate and safe complementary foods allow children in Western industrialized countries to eat more food safely and earlier than their ancestors or their contemporaries in Third World populations. Clearly advances in medical care, particularly immunizations against many childhood diseases, and antibiotics, make it possible for children in Western industrialized countries to be weaned relatively early without high mortality rates. Clearly advances in public and household sanitation—clean water, sewer systems, antibacterial soaps—make it possible for children in Western industrialized countries to avoid many bacterial, viral, and parasitic infections that still plague much of the world. Clearly, it is not necessary for mere survival that all children in Western industrialized countries are breast-fed until they are six years old, or even three or four years old.
At the same time, however, it is clear from the medical research that the longer a child is breastfed, the better the health outcome for that child, even under the best of First World conditions. Only a few studies have defined breast milk intake as a dose-response variable (Fredrickson, 1995), and looked at the effect of varying durations of breastfeeding on child health. Most studies define infant feeding as either breast or bottle, with the breastfeeding group getting breast milk for varying lengths of time. Some have used six-month interval categories, and examined health outcomes for children “not breastfed,” “breastfed for 0–6 months,” “breastfed for 6–12 months,” “breastfed for 12–18 months,” and “breastfed for 18–24 months or longer.” In every case, whatever the outcome health variable, when duration of breastfeeding was considered, the longer the child was breastfed, the lower the risk for the adverse health outcome. These results hold true for studies on malocclusion (Labbok and Hendershot, 1987), gastrointestinal disease (Howie et al., 1990), Sudden Infant Death Syndrome (Fredrickson et al., 1993; Mitchell et al., 1991), otitis media (Duncan et al., 1993), childhood cancer (Davis et al., 1988; Golding et al., 1990), juvenile diabetes (Cavallo et al., 1996; Dahl-Jørgensen et al., 1991; Mayer et al., 1988; Virtanen et al., 1991) and attention deficit hyperactivity disorder (Stevens et al., 1995). Likewise, the research on the relationship between cognitive achievement (IQ scores, grades in school) showed the greatest gains for those children breastfed the longest (Rogan and Gladen, 1993). To put it another way, the research showed that the shorter the duration of breastfeeding, the greater the cognitive deficits (Wiessinger, 1996).
Recent research also suggests that many health problems of adults can be traced to lack of breastfeeding or short durations of breastfeeding, suggesting that while children raised on formula may survive, they do not necessarily thrive as adults. Premature weaning has been identified as a factor in all of the following diseases or conditions of adulthood: inflammatory bowel disease (Acheson and Truelove, 1961), Crohn’s disease (Koletzko et al., 1989), celiac disease (Greco et al., 1988), Schönlein-Henoch purpura (Pisacane et al., 1992), breast cancer (Freudenheim et al., 1994), multiple sclerosis (Pisacane et al., 1994), allergies (many studies, reviewed in Cunningham, 1995), chronic respiratory diseases (also reviewed in Cunningham, 1995), and coronary artery disease (Fall et al., 1992; Kato et al., 1992; Kawasaki et al., 1974; Marmot et al., 1980; Osborn, 1968).
At this point in scientific understanding, no research is available that looks at the differentials in disease and cognitive outcomes for children breastfed longer than two years, compared with those breastfed for shorter periods. All of the studies that have examined the relationship between duration of breastfeeding and health outcomes stop with an upper category of 24+ months. No studies have compared the health differential between children breastfed for three years compared with those breastfed for four years, compared with those breastfed for five years, and so on. Thus, stating unequivocally that there are significant or substantial health benefits to breastfeeding beyond two years of age is not possible. At the same time, stating unequivocally that there are no significant or substantial health benefits to breastfeeding beyond two years of age likewise is not possible. The additional health benefits, in a First World setting, of breastfeeding beyond two years may be very slight, or they may be significant; they may affect health in childhood, or they may not become apparent until middle-age or later—the data are simply not available at this time.
Thus, health care professionals cannot guarantee mothers that continued breastfeeding will result in noticeable improvements in lifelong health or longevity for their children. What is more significant, and clear, is that health care professionals have no basis for claiming that the health benefits of breastfeeding ever cease or become insignificant, nor for claiming that extended breastfeeding ever harms a child or is contraindicated after a certain age. All of the available research shows better health outcomes the longer the child breastfeeds, up to the current study limits of 24 months. If the mother and child want to continue breastfeeding, no reasons exist to suggest that they should not. In particular, no reasons exist to question the motivations of the mother who wants to give her child the best possible lifelong health, nor to accuse her of harboring ulterior motives for continuing to breastfeed well beyond two years.
Undoubtedly, in terms of health outcomes, the first year of breastfeeding is more important than the fourth year. The first month is more important than the 24th month. The first week is more important than the 52nd week. “Any breastfeeding,” as a research category, has been shown to result in significant health improvements compared with children in the “no breastfeeding” category. No evidence exists to suggest that the health benefits ever cease, nor that the cost/benefit ratio ever becomes so high as to make continued breastfeeding pointless or contraindicated. Future research may confirm that breastfeeding for the full length of time normal for our species, in evolutionary and physiological terms, may result in children who thrive—in terms of physical health, cognitive development, and emotional stability—compared with children who were weaned before 2.5 years.
Prolonged breastfeeding and malnutrition/poor growth
A number of studies have suggested that prolonged breastfeeding may negatively affect children’s growth and nutritional status, as measured by anthropometric variables. Essentially, these studies use anthropometric measurements to compare the nutritional status of children who are still breastfeeding with the nutritional status of children who have been weaned at a particular age or age range. In some studies, the children who are still breastfeeding have lower nutritional status, as measured by z-scores for height-for-age and weight-for-height, than children of the same age who are already weaned (see Nubé and Asenso-Okyere, 1996). Some researchers have concluded, therefore, that there is a causal relationship between prolonged breastfeeding and poor nutritional status, and that the causal arrow goes from prolonged breastfeeding to poor nutritional status. Nubé and Asenso-Okyere (1996) conclude that prolonged breastfeeding is the cause of poor nutritional status in older children in Ghana who are still breastfeeding.
A critique of such studies and their conclusions was published by Grummer-Strawn in 1993, and a further study documenting the important role of breastfeeding in child survival was published by Mølbak et al. in 1994. Grummer-Strawn pointed out that in many cultures, the duration of breastfeeding depends in part on the health status of the child. Mothers recognize that breastfeeding promotes child survival and health, and sickly children are therefore nursed longer than healthy children, who can be safely weaned. The same rationale is often given for breastfeeding boys longer than girls in cultures where this is the practice; mothers recognize that boys tend to have higher morbidity and mortality than girls, so they nurse their sons longer (McKee, 1984). In my research in Mali, I found that prolonged breastfeeding to keep a sick child alive was one of the culturally recognized exceptions to the general limit of two years for breastfeeding (Dettwyler, 1987). Thus, a direct causal relationship between breastfeeding and malnutrition may sometimes exist, but the causal arrow goes in the opposite direction to that suggested by Nubé and Asenso-Okyere (1996) and other researchers. That is, evidence exists to explain how malnutrition and poor health lead to prolonged breastfeeding; there is no evidence that prolonged breastfeeding leads to malnutrition.
Another, critical failing of studies that report that “still breastfeeding” children have poorer nutritional status than “already weaned” children, is the researchers’ failure to control for mortality in their cross sectional samples. As expected in any Third World country such as Ghana, the site of Nubé and Asenso-Okyere’s study, childhood mortality (birth to 5 years of age) is quite high. In each age category, children with the poorest anthropometric scores will be those most at risk of dying from the synergistic actions of malnutrition, disease, and parasites. In the category of “children already weaned”—children who no longer have the protective immunological benefits of breast milk, nor the clean, complete protein, the calories, and the fat of breast milk—some proportion of those at the lower end of the range of anthropometric variation die. Thus, their very low z-scores are eliminated from their sub-sample, giving a false impression that their group’s nutritional status is relatively good. In the category of children “still breastfeeding”—children who continue to have the protective immunological benefits of breast milk, and the clean, complete protein, the calories, and the fat of breast milk—many of those at the lower end of the range of anthropometric variation survive. Thus, their very low z scores are included in their sub-sample, giving the impression that their group’s nutritional status is relatively poor. If some way could be developed to factor in a z-score for all the “completely weaned” children who had died because of being weaned, the apparent good nutritional status of this group would disappear.
In studies that collect data on mortality as well as nutritional status, one can see that the apparent better nutritional status of the already weaned group comes at the cost of many lives. This is clearly demonstrated in the study of breastfeeding, diarrhoeal disease and child mortality in Guinea Bissau carried out by Molbak and colleagues (Mølbak et al., 1994). They found that still-breastfeeding children in the 12–35 month age category did indeed have lower anthropometric scores than already weaned children of the same ages. However, they also found that the mortality rate among already weaned children was 3.5 times higher than the rate among still-breastfeeding children. In addition, the incidence of diarrhoea was higher and the duration of diarrhoea was longer, in the already weaned children compared to the still-breastfeeding children.
Thus, the conclusion that extended breastfeeding is correlated with poor anthropometric scores is true in a statistical sense; however, one major alternative to being breastfed and growing poorly is to be weaned and dead. Only those children whose health is very good—and whose good health may very well have contributed to their early weaning—will survive premature weaning.
This paradox, that breastfeeding and nutritional status, and nutritional status and mortality, are linked in ways that can seem counterintuitive, was first brought to my attention when I was conducting research on child feeding practices and growth in Mali in 1989 and 1990. Children in rural villages in southern Mali showed extremely high levels of malnutrition, peaking in the 2–3 year age range. The four year olds seemed to be greatly improved. Some of this improvement was due in part to a real recovery—they had survived many viral diseases that afflict young children, they were developing immunity to malaria, they were old enough to gather food for themselves, or to ask for more, and they were better able to compete with younger siblings for food from the communal bowl. Nevertheless, some of this improvement was also due to the deaths of the children who were most severely malnourished in the 2–3 year age range. They died, so they were not around to be measured as four year olds (Dettwyler, 1991).
This paradox was made even clearer when I worked with CARE in northern Mali. CARE had an extensive program that involved a series of health interventions in rural villages in the Macina region. They helped the villagers build concrete-lined and capped wells, to provide a clean water source for drinking and for irrigation of dry season gardens to increase the availability of garden vegetables (to combat Vitamin A deficiency, in part). They taught the villagers about household and village sanitation, especially the removal of goat manure and food debris that attracted flies and rodents. They taught birth hygiene classes to reduce the chances of complications during childbirth, including the elimination of the traditional practice of smearing cow dung on the umbilical stump, which caused a very high rate of death from neonatal tetanus. Perhaps most ambitiously, they provided immunizations against a variety of childhood diseases for all young children. I was brought in under a contract with the Academy for Educational Development (AED) to conduct interviews and focus group research in the villages, and conduct market food availability studies. Based on this research, I made recommendations for nutrition education messages that could be incorporated into a new nutrition component of the program that was being developed (Dettwyler 1989, 1990; Dettwyler and Fishman, 1990).
One thing that struck me immediately was the terrible nutritional status of the children in the CARE villages. At first, I thought it might be just that the northern villages were in much worse shape than the peri-urban community or the southern rural villages, where I had concentrated my previous and ongoing research. However, I also visited several rural villages in the Macina region that were not yet in the CARE program. Even without taking anthropometric measurements, I could see that the children in these villages had much better nutritional status than those in the CARE villages. My visual observations were later confirmed by an anthropometric survey carried out by AED; the CARE-village children were the most malnourished in the region. I described them as “Dancing Skeletons” (Dettwyler, 1994).
How was this possible? The explanation was quite simple. Most of the severely malnourished children in the CARE villages were still alive, while most of the severely malnourished children in the non CARE villages had died. Without the stress of viral diseases, and without the stress of diarrhea due to contaminated water, the malnourished children in the CARE villages were hanging on, they were surviving, though quite malnourished. Children of similar nutritional status in the non-CARE villages were pushed over the edge and died when they got measles, or diphtheria, or whooping cough, or they died when they got dysentery or giardia from contaminated water. A dead child cannot run out to greet the “toubab” (a white person) who comes to visit; a dead child cannot contribute his poor anthropometric measurements to the cross-sectional sample.
The idea that breast milk can suppress a child’s appetite for more nutritious foods (Nubé and Asenso Okyere, 1996) begs the question of what foods are better for the child, nutritionally, than breast milk? In the United States, some parents report that their child’s pediatrician urges them to wean so the child will eat more solid foods. Even in the United States, however, solid foods are not better nutritionally for the child. Neither commercially produced baby foods nor adult foods have higher nutritional value for a young child than breast milk. In the Third World context of Ghana or Mali, the alternatives to breast milk for young children are usually contaminated carbohydrate gruels. To imply that a child will be healthier if the one source of clean, complete protein is removed from his or her diet makes no sense at all. The conclusions reached by Nubé and Asenso-Okyere are not supported by an informed interpretation of their own data, and reflect a misunderstanding of the sad reality of high childhood mortality in Third World populations. To blame breastfeeding, the one factor keeping these children alive, for their poor nutritional status, is wrong. To recommend that mothers limit their children’s access to breast milk is misguided and dangerous. No data, from the perspectives of health, nutritional status, or growth, warrant the conclusion that mothers should wean their children at one or two years of age, or that international health organizations should establish recommended upper limits on duration of breastfeeding.
Implication for clinical practice: “Should we advocate nursing them til they are six?”
In helping a mother to decide how long to continue breastfeeding, health care professionals are obligated ethically to present the current understanding of the continued benefits and the current understanding of the normal, physiological duration of breastfeeding for the species, and then let the mother decide. The importance of the evolutionary perspective is not to recommend that all mothers should nurse their children for a particular length of time. The importance of this perspective is to de-pathologize species-normal breastfeeding and weaning patterns that are still practiced by many mothers and children today, whether in a remote village in Mali, or in suburban middle-class America.
Some people have argued that the reduction in risk of illness provided by extended breastfeeding is not sufficient to warrant the behavior: “Is it worth changing one’s behavior—breastfeeding instead of bottle-feeding or breastfeeding for four years instead of two years—to reduce the risk of a rare event?” What is clear from the research comparing the health of breastfed and bottle-fed children is that, while the reduction in risk for some diseases may be small, and while some diseases, such as childhood lymphoma, are very rare to begin with, other reductions in risk are very large (exclusive breastfeeding to six months reduces the risk of Sudden Infant Death Syndrome almost to zero, Fredrickson et al., 1993), and some diseases/conditions, such as asthma and allergies, are very common in formula-fed infants. Overall, the accumulated reduction in risk across all the diseases studied to date is substantial. Many researchers think that the lower risk of many diseases in breastfed infants is due to their stronger immune systems, rather than specific mechanisms of protection for each condition. Thus, lower rates of lymphoma and multiple sclerosis in breastfed infants probably can be attributed to successful defense of the body by a strong immune system. New or different diseases may evolve that will affect today’s children when they are adults, diseases for which a strong immune system may be the best, or the only, defense.
In choosing whether, and for how long, to breastfeed, parents are making decisions that will have long lasting consequences for their children’s health. Parents need to be fully informed about the risks that may be avoided by breastfeeding, to make informed and responsible decisions. Just as parents today are informed about the risks involved in not using child safety seats, and drinking during pregnancy—dangers unknown a generation ago—parents today need access to information about the risks of infant formula and of a culturally-prescribed shortened duration of breastfeeding (Walker, 1998; Wiessinger, 1996).
Every mother should be encouraged and supported to breastfeed her child for 2.5 years or longer. This includes a willingness, by health care providers, to speak out about the physiological normalcy of breastfeeding the older child. For the many mother-child pairs who are currently breastfeeding in secret, for fear of social censure, we must de-pathologize ‘extended’ breastfeeding (Kendall-Tackett and Sugarman, 1995; Sugarman and Kendall-Tackett, 1995). Family court lawyers, judges, teachers, social workers, friends and family, all need to recognize that the mother who is breastfeeding a child beyond toddlerhood is acting in a way that is healthy for both child and mother, physiologically normal and evolutionarily adaptive.
For mothers who decide to wean prematurely, health care professionals must provide education and support to help relieve the negative health consequences. This might include discussion of various types of formula, and advising parents to avoid whole cows’ milk until the child is at least 2.5 years old. It might include making sure all immunizations are complete and up-to-date, and that parents understand the necessity for prompt treatment of all illnesses. Dietary supplements of docosahexanoic acid and arachidonic acid might be advised for optimal brain growth in the child who has been weaned before brain growth has been completed. Parents should also be educated about the possibility that their child’s instinctive need to suck (which can be very strong in some children up to and beyond seven years of age) should be respected and met in another way, if not at the breast. Thus, pacifier use and thumb-sucking through age seven, at least, should be considered normal behavior for prematurely weaned children. While pacifiers and thumb-sucking are less healthy alternatives for meeting a child’s sucking needs than at the breast, they must still be recognized as meeting legitimate needs. If thumb sucking is leading to orthodontic problems, then alternatives need to be found. Never should the child, or the parents, be criticized for behavior that is normal for the circumstances and meets real needs. “Extended breastfeeding” should not be viewed as pathological or evidence of sexual abuse, or even of an overly-attached mother-child pair; it should be viewed as normal. Finally, education and support from health care officials will be necessary to counteract any negative health consequences to the mother of curtailed lactation, such as increased risk of breast cancer and osteoporosis, and a need for birth control.
Implications for research: The terrains of ignorance surrounding premature weaning
Very little research is currently available that takes into account this new realization that our species has evolved to expect many years of breastfeeding, and many years of having breast milk as a part of the child’s diet. Suggestions for research opened by this evolutionary perspective might include the following topics:
Basic research on extended lactation Research is needed on normal breast milk production beyond two years postpartum. All currently available studies of lactation end at the two-year mark. We need research on the quantity of milk produced under different circumstances, the composition of milk beyond two years postpartum, and on the duration of time lactation is possible following one pregnancy.
Health outcomes for children We need analyses of health outcomes for children breastfeeding beyond two years of age, using breast milk intake as a dose-response variable. Longitudinal follow-up studies, into later adulthood and old age, of children breastfed 3, 4, 5, 6, and seven years are necessary to find out the long-term health consequences of breastfeeding for different durations. For all of the diseases and conditions currently known to be reduced in children who breastfeed for up to two years, are they reduced even further in children who breastfeed for additional years? Research suggests that breastfeeding reduces the need for later orthodontic treatment, but none of the studies done to date have looked at extended breastfeeding in terms of development of the oral cavity. Do children who breastfeed for several years have “normal” development of the oral cavity? Do they need less orthodontic treatment than children breastfed for one to two years? Do they have less sleep apnea as adults, and less need for speech therapy (see for example the work of Palmer 1998, 1999)? Do children who breastfeed for many years grow up to be healthier adults, with lower risks for late-onset diseases and conditions? Do children need extended breastfeeding for normal/optimal development of their immune systems?
Cognitive outcomes for children Analyses of cognitive outcomes are needed for children breastfeeding beyond two years of age, using breast milk intake as a dose-response variable. Do the steadily increasing cognitive scores of breastfed children up to two years of breastfeeding continue in the years beyond, or do they reach a plateau after a certain age? Are there specific aspects of cognitive development affected more by duration of breastfeeding than others? Can we separate the influence of the product, breast milk, from the influence of the process, breastfeeding, in longer durations of breastfeeding?
Psychological outcomes for children Research on the psychological consequences of breastfeeding for many years is entirely lacking. Although some psychologists view ‘extended breastfeeding’ as unusual and abnormal, and will argue that it is psychologically harmful to children and a symptom of ‘enmeshment’ of the mother and child (Love, 1991), no research data exist to support these views. At the same time, anecdotal reports from parents practicing ‘Attachment Parenting’ (extended breastfeeding with child-led weaning, co-sleeping, lots of physical and emotional affection) suggest that these children grow up happy, independent, and well adjusted. Parents commonly report that they are asked, “What is your parenting secret?” However, no double-blind case-control studies have been conducted to document either negative or beneficial psychological consequences of extended breastfeeding, in either the short- or long-term. Random assignment to breastfeeding or bottle-feeding categories would be difficult to implement, and unethical, given the known dangers of infant formula. However, longitudinal research on matched case-control pairs could be done.
The evolutionary perspective suggests that normal psychological development in humans occurs under conditions of extended breastfeeding, extensive physical contact between mother and child in the early years, and routine co-sleeping. However, most of what we think we know about “normal early childhood development” is based on research done with children who were bottle/formula-fed, or breastfed according to a schedule for less than one year, and who routinely slept apart from their parents. Standard, basic child development research done with populations of children who breastfed, on cue, for 2.5 years or longer and coslept with their parents during childhood might reveal that many features of “normal child development” are adaptations or accommodations by children to premature weaning and lack of sufficient physical contact (sensory deprivation) during the early childhood years.
For example, many psychologists consider the development of an intense relationship with a soft object such as a doll or blanket to be a normal part of toddlerhood. Such a relationship with an inanimate object is called a “transitional object attachment,” and is thought to be not only normal, but healthy and necessary for the development of the child’s sense of self (Cohen and Clark, 1984). Children who do not develop such attachments are suspected of being abnormal. However, Cohen and Clark, citing Gaddini and Gaddini (1970) and Hong and Townes (1976), admit that cross-cultural studies reveal that “factors such as breast feeding, weaning at a later age, and sleeping in close proximity to the mother” are related to lower frequencies of transitional object attachments (1984:107). Likewise, anecdotal reports suggest that most children in the United States who are breastfed on demand for many years and co-sleep with their parents form their primary attachments to people, and do not usually develop intense emotional attachments to inanimate objects. It could be that the development of strong affective ties to an inanimate object reflect attempts by children to cope with the emotional bleakness of premature weaning, solitary sleeping, and lack of physical contact. Research on the prevalence of transitional object attachments among children in Western industrialized countries who breastfeed for more than 2.5 years and routinely cosleep with their parents would clarify whether such transitional object attachments are “normal and healthy” or are accommodations to premature weaning.
Some psychologists warn of the dangers of encouraging children to be securely attached to their mothers through extended breastfeeding and co-sleeping, arguing that this makes it more difficult for the child to develop into an independent adult (see for example Rath and Okum, 1995; see also the response by Dettwyler, 1997). One pop psychology book even includes extended breastfeeding and co-sleeping on its list of “the warning signs of emotional incest,” without a single citation of scientific research to justify doing so (Love, 1991). On the contrary, anecdotal evidence from parents suggests that children whose emotional needs are met in early childhood grow into confident, independent older children and adults. However, as with so many other issues, no well-designed studies have been conducted documenting the results of an attachment style of parenting, including several to many years of breastfeeding and co-sleeping, on later childhood and adult emotional health.
Additionally, many American mothers report that children who are still breastfeeding typically do not suffer from the emotional outbursts and frustrations commonly known as the “terrible twos.” Research is needed on how extended breastfeeding may help a young child cope with developmental milestones such as learning to walk and run, developing social relationships with other children, and dealing with the many frustrations of the toddler and post-toddler stages of life. If human children have evolved to expect to have the safe haven of mother’s arms and breasts to retreat to in times of fear, pain, frustration, and exhaustion, and they are deprived of this through premature weaning, might tantrums be the response? An illuminating example comes from Jane Goodall’s work, as reported by Frans de Waal, which is worth quoting in its entirety: “The mother has weapons, too [in the compromise over weaning]. Goodall (personal communication) offers a striking example with regards to the interbirth intervals of Fifi at Gombe National Park. Fifi had had regular 5-year intervals, almost like clockwork. Faustino was the first offspring born after a shorter interval: 4.5 years after the previous birth. This was only a minor deviation, however. She next gave birth in 1992, when Faustino was only 3.5 years old. The abrupt drop in maternal attention and opportunities for nursing caused unprecedented tantrums in Faustino. To him, life definitely must have seemed ‘unjust’. Fifi did not completely wean him, however, but continued to allow him to sleep in her nest, and even to nurse along with his new sibling. One of Fifi’s answers to Faustino’s tantrums was to climb high up in a tree and throw him to the ground, while at the last instant holding on to an ankle. The young male would hang upside down for 15 seconds or more, screaming his head off, before his mother would retrieve him. Goodall saw Fifi employ this scare tactic twice in a row, after which Faustino stopped having tantrums that day.” (de Waal, 1996:164–165).
Another ‘terrain of ignorance’ is the degree to which premature weaning (or not being breastfed at all) contributes to such clinical entities as attention deficit disorder, attention deficit hyperactivity disorder, autism/pervasive developmental disorder, and other cognitive/behavioral disorders in children. Does bottle-feeding contribute to the etiology of these conditions, or make the symptoms more severe? To phrase it the other way, does breastfeeding (or extended breastfeeding) reduce the risk of these conditions and/or lessen the severity of symptoms? Two anecdotal reports by mothers suggest that autistic children who had been breastfed for many years were more open to physical contact with other people. However, a search of the medical literature back to 1966 revealed only two studies of autism/pervasive developmental disorder (PDD) that included infant feeding as a variable (Burd et al., 1988; Tanoue and Oda, 1989).
Burd et al. (1988) looked at breastfeeding rates among children with PDD, matched controls, and their older and younger siblings, to see if PDD affected the frequency or duration of breastfeeding in the cases or their siblings. Tanoue and Oda (1989) found that their controls had longer durations of breastfeeding than their cases (children who developed autism), even though the children in their sample were only breastfed for a very short time. They suggest that since pneumonia in early infancy is a known precipitating factor for autism, and since breastfeeding protects against pneumonia, this may be the mechanism through which breastfeeding lowers the risk of developing autism. No studies have looked at the severity of autistic symptoms based on duration of breastfeeding. In particular, no studies have looked at cases of autism thought to be of prenatal origin (due to mother’s illness during pregnancy) and later cognitive and behavioral outcomes based on duration of breastfeeding.
Only one study of attention deficit hyperactivity disorder (ADHD) has looked at the role of infant feeding in the etiology and severity of this condition. Stevens et al. (1995) conducted a case-control study and found that children with ADHD seemed to have deficiencies of essential fatty acids, perhaps related to abnormal fatty acid metabolism. Their control group had a higher frequency of breastfeeding (81.4%) compared with the ADHD group (43.4%). Additionally, among those children in each group who were breastfed, the control group had a higher average duration (6.5 months) compared with the ADHD group (2.5 months). This study raises the intriguing possibility that the absence of long-chain polyunsaturated fatty acids in commercial infant formula may contribute to abnormal fatty acid metabolism in susceptible children who are not breastfed or who are weaned very early. It would be particularly interesting to study the prevalence and severity of ADHD in children who were breastfed for 2.5 years or longer.
In terms of long-term psychological outcomes, James Prescott has hypothesized, based on decades of research on nonhuman primates, that lack of affectional bonding during childhood, including being held and rocked, and breastfeeding for at least two years, may be a significant contributor to depression, impulse dyscontrol, and violence in adults. Prescott’s research suggests that the lack of breast milk and breastfeeding during the first two years of life may disrupt normal brain development, including serotonin-regulating mechanisms (Prescott 1990, 1996). Abnormal serotonin metabolism has been implicated in many mental illnesses, especially depression. Research is currently underway to test these theories on adults who were breastfed for at least two years (Prescott, pers. comm.).
Health outcomes for women We need a better understanding of the health consequences for women of breastfeeding for many years, for only a few years, or of not breastfeeding at all. Women’s breasts are often discussed in medical circles only in terms of their potential for breast cancer; however, many studies have found that breastfeeding lowers a woman’s risk of breast cancer (reviewed in Micozzi, 1995). Recent research by Newcomb et al. (1994) and Romieu et al. (1996) confirms these earlier findings. Newcomb et al. (1994), studying women in the United States, found that the longer a woman breastfed (summing up her total months of lactation across all her children), up to the study limits of a category labeled “24+ months,” the lower her risk of premenopausal breast cancer compared with women who had never lactated. They conclude that if parous women who never breastfed, or who breastfed for less than three months, were to do so for 4–12 months, breast cancer among parous premenopausal women could be reduced by 11 percent; if all parous women lactated for 24 months or longer, the incidence might be reduced by nearly 25 percent. In a study in Mexico, Romieu et al. (1996) found that parous women who had lactated had a 39 percent reduction in breast cancer risk compared with parous women who had never lactated. Those who breastfed for 12–24 months had a 47 percent reduction in breast cancer risk. Both pre- and post-menopausal breast cancer risk were reduced with longer durations of lactation after the second live birth. What might a study show that examined breast cancer risk in women who nursed for a total of five years, ten years, fifteen years, or more?
Only one study has examined whether having been breastfed lowers a woman’s risk of breast cancer when she grows up. This study, by Freudenheim et al. (1994) involved women from New York. “Breastfeeding” was defined as any breastfeeding, so some of these women may have only been breastfed for a week or a month, while others may have been breastfed for several years. The researchers found that, for both premenopausal and postmenopausal breast cancer, women who were breastfed as children, even if only for a short time, had a 25% lower risk of developing breast cancer than women who were bottle-fed as infants. What might a study show that examined breast cancer rates in women who had themselves been breastfed for two years, three years, four years or longer?
Romieu et al. conclude their study by pointing out that “The declining trend in fertility and lactation among Mexican women could lead to a major epidemic of breast cancer such as that observed in Western countries” (1996:543). Does the trend toward shortened durations of breastfeeding help account for the rise in breast cancer rates among women in Western, industrialized countries in this century? Most women born in the early decades of the century were breastfed as children, but then used bottles and formula for their children born in the 1940s and 1950s. These women had protection from having been breastfed themselves, but gave up the protection that comes from breastfeeding their own children. Their children, specifically the daughters born in the 1940s and 1950s, in turn mostly used bottles and formula when they were having children in the 1960s, 1970s, and 1980s. Thus, many women born in the 1940s and 1950s had neither form of protection. These women, now in their 40s and 50s, have skyrocketing rates of breast cancer—is there a causal connection? Research looking at the feeding histories and cancer rates of women across the decades of the 20th century, with significant numbers of women who were themselves nursed for several years, and/or who nursed several children for several years each, might confirm or refute this speculation.
It is known that breastfeeding results in a net gain in bone mineral density and a reduction in osteoporosis and hip fractures in later life (Blaauw, et al., 1994, Cumming, et al., 1993, Kalkwarf et al., 1995, 1996, Sowers et al., 1995). As with so many other diseases and conditions, however, all of the studies to date have been conducted on women who breastfed for relatively short durations. Many other potential health outcomes for women might be addressed through studies of extended breastfeeding.
Psychological outcomes for women The physical act of breastfeeding a baby releases two hormones, oxytocin and prolactin, which have profound effects on maternal feelings and behaviors. Both oxytocin and prolactin help the mother cope with the demands and anxieties of caring for a newborn. Much is known about the role of these two ‘mothering hormones’ in the early months postpartum, but to date, no research has been conducted on how these hormones might help a mother cope with a toddler or older child. I suggested earlier that tantrums might be less frequent in the older breastfeeding child; a mother’s ability to cope with difficult behaviors from an older child logically would also be affected by her hormonal state. Additionally, one study suggests that lack of breastfeeding and/or premature weaning may contribute significantly to postpartum depression (Peyton, 1996). From an evolutionary perspective, lack of any suckling, or early cessation of suckling, signals to the mother’s body that the child has died. Her hormonal state reverts to that of a mother without a dependent offspring, and her body gears up to try again for successful reproduction. The mismatch between her body’s hormonal state, with its very low levels of the two appropriate mothering hormones, oxytocin and prolactin, and the reality of having a young infant to care for—an infant who may be crying a lot, experiencing physical and emotional distress from not being breastfed, and who is sick more often because of not being breastfed—may be enough to plunge a new mother into postpartum depression.
Many psychologists and health care professionals have defined the non-pregnant, non-lactating hormonal state as “normal,” when clearly, from an evolutionary perspective, this hormonal state is not at all normal for a new mother. High levels of oxytocin and prolactin due to breastfeeding are the “normal” hormonal state of a mother with a young child. Indeed, in evolutionary terms, high levels of these hormones due to breastfeeding are the normal hormonal state of women throughout their reproductive years, except for the intervals of pregnancy. Postpartum depression is most often studied in the newborn period, but premature weaning before the species-normal minimum of 2.5 years postpartum may place women at risk of depression and other difficulties, when a mother without the benefits of the mothering hormones must cope with a young child who is without the calming and comforting benefits of breastfeeding.
Research on how to mimic the beneficial effects of extended breastfeeding for those women who do not experience it If it were widely recognized that many years of breastfeeding was the species-normal pattern, and if extended breastfeeding were supported and encouraged, many women in Western, industrialized countries would feel more confident about breastfeeding until their children weaned themselves. However, most women in these populations will still have only one or two children during their reproductive careers, and there will still be many women who do not breastfeed their children for very long, or at all. Therefore, research is needed to develop other means of replacing the protection that women used to get from many years of breastfeeding. Can hormonal treatments be developed that mimic the benefits of prolonged lactation, in terms of breast cancer, osteoporosis, and other factors, without the side-effect of milk production? Can hormonal treatments be developed that mimic the mothering hormones, as treatment for postpartum depression where breastfeeding is impossible, is not chosen, or is cut short?
Research on how to mimic the beneficial effects of extended breastfeeding for those children who do not experience it Can medical treatments be developed that will help strengthen the immune system for adults who were never breastfed, or who were weaned prematurely? Can formulas be developed that include docosahexanoic acid and arachidonic acid for better cognitive development, without having negative side-effects on infant physical growth? Can mothers be taught how to bottle-feed in ways that approximate the close physical, social, and affectional interactions that occur naturally during breastfeeding?
Research on how to best meet the sucking needs of children weaned prematurely There are many questions that could be asked on this topic: Is a pacifier preferable to the child’s own thumb or fingers? If so, what shape, size, and type of pacifier would be best? Is there a certain technique of thumb- or finger-sucking that results in fewer orthodontic problems? Is orthodontic treatment during the teenage years a reasonable trade-off for allowing the child to suck its thumb as long as it needs to during childhood?
Research on the cultural causes and demographic consequences of reducing the natural duration of breastfeeding Research to examine the demographic consequences of reducing the natural duration of breastfeeding through cultural beliefs would be very useful. Was the increase in population that accompanied the domestication of plants and animals related to shorter interbirth intervals due to earlier weaning? Was earlier weaning possible, in an evolutionary sense, because of the domestication of grain crops from which cereal porridges could be made, or does it go back even further? Is it related to the discovery of how to control and use fire to cook food? Or is it more closely related to the cultural beginnings of medical care? Did a cultural pattern of early weaning allow early human populations to reduce the interbirth interval, and have more offspring, who still had a good chance of survival? Does this explain the emergence of humans as the most numerous and widespread primate species? Does it explain, in part, recent dramatic increases in world population?
Women in most industrialized countries today follow an extremely k-selected pattern of having relatively few children, in whom they invest heavily with their time, effort, and money. The typical Western pattern of shortening the breastfeeding period as much as possible and having children close together in age—and then spending most of one’s reproductive career artificially preventing further pregnancies—makes little evolutionary, biological, or even cultural sense. In such a context, a more rational strategy would be to return to a pattern of maximizing the breastfeeding investment in each offspring, thereby maximizing the health outcome and cognitive achievement of each child. This requires having those offspring relatively far apart, or tandem nursing two children of different ages (common among women in the U.S. practicing child-led weaning). How might U.S. cultural values be influenced to encourage parents to space their children farther apart and to breastfeed each one for at least 2.5 years?
Cultural research on the reasons why extended breastfeeding came to be viewed as abnormal or unnatural in Western industrialized countries, and how to change such perceptions From whence comes the cultural belief system that defines extended breastfeeding as problematic? Is it due entirely to the sexualization of women’s breasts in the United States and other Western cultures (Dettwyler, 1995b)? Or is it related as well to more general beliefs about the importance of independence in U.S. culture, and the general trend toward minimizing strong affective bonds between parents and children through such practices as separation of mother and newborn in the hospital after birth, and later separate sleeping arrangements in the home (culminating in the ‘split bedroom plan’ of many new homes, where the master bedroom suite is found on the opposite side of the house from the children’s rooms, or even on a different floor)? What role do parental convenience and desire for privacy play in this? What role is played by a focus on the conjugal bond in Western industrial countries, rather than the mother-child bond typical of nonhuman primates, and still predominant in many ‘traditional’ cultures worldwide? To what extent has research been hindered by the ethnocentric attitudes of some researchers, who define their own, culturally sanctioned child-rearing practices as automatically normal and best for children’s health, and/or by researchers who view extended breastfeeding as something only ‘primitive’ people do? To what extent has research been hindered by a nonevolutionary approach that fails to recognize humans as primate mammals, and therefore defines extended breastfeeding as abnormal? To what extent has research been hindered by feelings of guilt and defensiveness among researchers whose children were not breastfed, or not breastfed for very long?
Research on the consequences today for Third World populations who are moving toward earlier and earlier weaning Research is needed not just on the health consequences for the children, which are already quite well documented, but on the health consequences for the mothers, who are having many more pregnancies and births than they did traditionally. To what extent does premature weaning contribute to explosive population growth and maternal mortality around the world?
CONCLUSIONS
The evolutionary perspective on human life history variables suggests that, like all other segments of the life span, duration of breastfeeding in humans would naturally be lengthened relative to the other primates. Just as gestation takes 38 weeks in humans compared with 33 weeks in the chimpanzee, so breastfeeding would be expected to last in humans at least as long as the 4.5 to 7.0 years of breastfeeding reported for chimpanzees (van Lawick-Goodall, 1973). Analysis of the nonhuman primate data on life history variables provides a clear picture of the species-normal duration of breastfeeding in modern humans. Extended breastfeeding, from a minimum of 2.5 years, to a maximum of 7+ years, is healthy, physiologically normal and evolutionarily adaptive. It meets the needs of children for nutritional, immunological, and emotional sustenance until they are ready and able to meet those needs on their own. Duration of breastfeeding is the only life span stage amenable to direct and substantial cultural manipulation, and the only one for which routine shortening or complete elimination has become accepted as the cultural norm. It is the only life history variable for which maternal behaviors that match evolutionary expectations have been defined culturally as abnormal and suspect.
The recognition that 2.5 to seven years of breastfeeding is what modern human children have evolved to expect casts serious doubts on research suggesting that prolonged breastfeeding is detrimental to children under any but the most unusual of circumstances (such as HIV infection of the mother). It also opens many new areas for research into the consequences of limiting the duration of breastfeeding on child physical health, emotional health, and cognitive development, and maternal physical and emotional health. Such recognition also provides a different perspective on behaviors commonly seen in bottle-fed children such as thumb-sucking and attachment to inanimate objects, and suggests that we know very little about normal child development under the evolutionarily-expected conditions of extended breastfeeding.
REFERENCES
Acheson, E.D., and Truelove, S.C. (1961) Early weaning in the aetiology of ulcerative colitis. British Medical Journal 4:929–933.
Blaauw, R., Albertse, E.C., Beneke, T., Lombard, C.J., Laubscher, R., and Hough, F.S. (1994) Risk factors for development of osteoporosis in a South African population. South African Medical Journal 84:328–332.
Burd, L., Fisher, W., Kerbeshian, J., Vesely, B., Durgin, B. and Reep, P. (1988) A comparison of breastfeeding rates among children with pervasive developmental disorder, and controls. Developmental and Behavioral Pediatrics 9:247–251.
Cabana, T., Jolicouer, P., and Michaud, J. (1993) Prenatal and postnatal growth and allometry of stature, head circumference, and brain weight in Québec children. American Journal of Human Biology 5:93–99.
Cavallo, M.G., Fava, D., Monetini, L., Barone, F. and Pozzilli, P. (1996) Cell-mediated immune response to ß casein in recent-onset insulin-dependent diabetes: Implications for disease pathogenesis. Lancet 348:926–928.
Charnov, E.L. (1991) Evolution of life history variation among female mammals. Proc. Natl. Acad. Sci. 88:1134–1137.
Charnov, E.L. (1993) Life history Invariants. Oxford: Oxford University Press.
Charnov, E.L. and Berrigan, D. (1993) Why do female primates have such long life spans and so few babies? or Life in the slow lane. Evolutionary Anthropology 1:191–194.
Cohen, K.N., and Clark, J.A. (1984) Transitional object attachments in early childhood and personality characteristics in later life. Journal of Personality and Social Psychology 46:106–111.
Cumming, R.G., and Klineberg, R.J. (1993) Breastfeeding and other reproductive factors and the risk of hip fractures in elderly women. Int. J. Epidemiol. 22(4):684–691.
Cunningham, A.S. (1995) Breastfeeding: Adaptive behavior for child health and longevity. In Breastfeeding, Biocultural Perspectives, ed. P. Stuart-Macadam and K. A. Dettwyler, pp. 243–264. New York: Aldine de Gruyter.
Dahl-Jørgensen, K., Joner, G. and Hanssen, K.F. (1991) Relationship between cow’s milk consumption and incidence of IDDM in childhood. Diabetes Care 14:1081–1083.
Davis, M.K., Savitz, D.A., and Graubard, B.I. (1988) Infant feeding and childhood cancer. Lancet 2:365–368.
de Andraca, I., and Uauy, R. (1995) Breastfeeding for optimal mental development: The alpha and the omega in human milk. In Behavioral and Metabolic Aspects of Breastfeeding. World Review of Nutrition and Dietetics, 78, ed. A.P. Simopoulos, J.E. Dutra de Oliveira, and I.D. Desai, pp. 1–27. Basel: Karger.
Dettwyler, K.A. (1985) Breastfeeding, Weaning, and Other Infant Feeding Practices in Mali and Their Effects on Growth and Development. Ph.D. dissertation, Department of Anthropology, Indiana University, Bloomington, Indiana.
Dettwyler, K.A. (1987) Breastfeeding and weaning in Mali: Cultural context and hard data. Social Science and Medicine 24:633–644.
Dettwyler, K.A. (1989) Field Research in Macina for Vitamin A Communications, 28 November 5 December 1989. Report prepared for the Nutrition Communication Project, Communication for Vitamin A, Academy for Educational Development (in cooperation with the U.S. Agency for International Development and CARE/Mali).
Dettwyler, K.A. (1990) Where There is No Anthropometer: A Field Guide for Identifying Children with Malnutrition and Tips for Counseling their Parents. Training Module prepared for the Academy for Educational Development (in cooperation with the U.S. Agency for International Development and the Ministry of Health, Republic of Mali).
Dettwyler, K.A. (1991) Growth status of children in rural Mali: Implications for nutrition education programs. American Journal of Human Biology 3:447–462.
Dettwyler, K.A. (1992) Nutritional status of adults in rural Mali. American Journal of Human Biology 88:309–321.
Dettwyler, K.A. (1994) Dancing Skeletons: Life and Death in West Africa. Prospect Heights, IL: Waveland Press.
Dettwyler, K.A. (1995a) A time to wean: The hominid blueprint for the natural age of weaning in modern human populations. In Breastfeeding: Biocultural Perspectives, ed. P. Stuart-Macadam and K.A. Dettwyler, pp. 39–73. New York: Aldine de Gruyter.
Dettwyler, K.A. (1995b) Beauty and the breast: The cultural context of breastfeeding in the United States. In Breastfeeding: Biocultural Perspectives, ed. P. Stuart-Macadam and K.A. Dettwyler, pp. 167–215. New York: Aldine de Gruyter.
Dettwyler, K.A. (1997) On parents and children sleeping together (letter). American Journal of Orthopsychiatry 67:159–160.
Dettwyler, K.A., and Fishman, C. (1990) Field Research in Macina for Vitamin A Communications, March 5–22, 1990. Report prepared for the Nutrition Communication Project, Communication for Vitamin A, Academy for Educational Development (in cooperation with the U.S. Agency for International Development and CARE/Mali).
de Waal, F.B.M. (1996) Conflict as negotiation. In Great Ape Societies, ed. W. C. McGrew, L. F. Marchant, and T. Nishida, pp. 159–172. Cambridge: Cambridge University Press.
Duncan, B., Ey, J., Golberg, C.J., Wright, A.L., Martinez, F.D., and Taussig, L.M. (1993) Exclusive breast-feeding for at least 4 months protects against otitis media. Pediatrics 91:867–872.
Eveleth, P.B., and Tanner, J.M. (1990) Worldwide Variation in Human Growth. Cambridge: Cambridge University Press.
Fall, C.H., Barker, D.J., Osmond, C., Winter, P.D., Clark, P.M., and Hales, C.N. (1992) Relation of infant feeding to adult serum cholesterol concentration and death from ischaemic heart disease. British Medical Journal 304:801–805.
Fredrickson, D.D. (1995) Commentary: Breastfeeding study design problems—health policy, epidemiologic and pediatric perspectives. In Breastfeeding: Biocultural Perspectives, ed. P. Stuart-Macadam and K.A. Dettwyler, pp. 405–418. New York: Aldine de Gruyter.
Fredrickson, D.D., Sorenson, J.R., Biddle, A.K., and Kotelchuck, M. (1993) Relationship between sudden infant death syndrome and breastfeeding intensity and duration. American Journal of Diseases of Children 147:460.
Freudenheim, J.L., Marshall, J.R., Graham, S., Laughlin, R., Vena, J.E., Bandera, E., Muti, P., Swanson, M., and Nemoto, T. (1994) Exposure to breast milk in infancy and the risk of breast cancer. Epidemiology 5:324–331.
Gaddini, R., and Gaddini, E. (1970) Transitional objects and the process of individuation: A study in three different social groups. Journal of the American Academy of Child Psychiatry 9:347–365.
Garn, S.M., and Bailey, S.M. (1978) Genetics of maturational processes. In Human Growth 1, Principles and Prenatal Growth, ed. F. Falkner and J.M. Tanner, pp. 307–330. New York: Plenum Press.
Ginsburg, H., and S. Opper (1979) Piaget’s Theory of Intellectual Development. Englewood Cliffs, NJ.: Prentice-Hall, Inc.
Golding, J., Paterson, M., and Kinlen, L.J. (1990) Factors associated with childhood cancer in a national cohort study. British Journal of Cancer 62:304–308.
Graham, C.E., ed. (1981) Reproductive Biology of the Great Apes. New York: Academic Press.
Greco, L., Auricchio, S., Mayer, M., and Grimaldi, M. (1988) Case-control study on nutritional risk factors in celiac disease. Journal of Pediatric Gastroenterology and Nutrition 7:395–399.
Grummer-Strawn, L.M. (1993) Does prolonged breastfeeding impair child growth? Pediatrics 91:766–771.
Hahn-Zoric, M., Fulconis, F., Minoli, I., Moro, G., Carlsson, B., Böttiger, M., Räihä, N., and Hanson, L.Å. (1990) Antibody responses to parenteral and oral vaccines are impaired by conventional and low protein formulas as compared to breast-feeding. Acta Paediatrica Scandinavica 79:1137–1142.
Hamill, P.V.V., Drizd, T.A., Johnson, C.L., Reed, R.B., Roche, A.F., and Moore, W.M. (1979) Physical growth: National Center for Health Statistics percentiles. American Journal of Clinical Nutrition 32:607–629.
Harvey, P.H., and Clutton-Brock, T.H. (1985) Life history variation in primates. Evolution 39:559–581.
Hong, K., and Townes, B. (1976) Infants’ attachment to inanimate objects. Journal of the American Academy of Child Psychiatry 15:49–61.
Howie, P.W., Forsyth, J.S., Ogston, S.A., Clark, A., and Florey, C.D. (1990) Protective effect of breast feeding against infection. British Medical Journal 300:11–16.
Jamison, P.L. (1978) Anthropometric variation. In Eskimos of Northwestern Alaska: A Biological Perspective, ed. P.L. Jamison, S.L. Zegura, and F.A. Milan, pp. 40–78. US/IBP Synthesis Series No. 8. Stroudsburg, PA.: Dowden, Hutchinson & Ross, Inc.
Johnson, P.L., Wood, J.W., Campbell, K.L., and Maslar, I.A. (1987) Long ovarian cycles in women of highland Papua New Guinea. Human Biology 59:837–845.
Jolicouer, P., Baron, G., and Cabana, T. (1988) Cross-sectional growth and decline of human stature and brain-weight in 19th Century Germany. Growth, Development, and Aging 52:201–206.
Kalkwarf, H.J., Specker, B.L., Heubi, J.E., Vieira, N.E., and Yergey, A.L. (1996) Intestinal calcium absorption of women during lactation and after weaning. Amer. J. Clin. Nutr. 63(4):526–531.
Kato, H., Inoue, O., Kawasaki, T., Fujiwara, H., Watanabe, T., and Toshima, H. (1992) Adult coronary artery disease due to childhood Kawasaki disease. Lancet 340:1127–1129.
Kawasaki, T., Kosaki, F., Okawa, S., Shigematsu, I., and Yanagawa, H. (1974) A new infantile acute febrile mucocutaneous lymph node syndrome (MLNS) prevailing in Japan. Pediatrics 54:271–281.
Kendall-Tackett, K.A., and Sugarman, M. (1995) The social consequences of long-term breastfeeding. Journal of Human Lactation 11:179–183.
Koletzko, S., Sherman, P., Corey, M., Griffiths, A., and Smith, C. (1989) Role of infant feeding practices in development of Crohn’s disease in childhood. British Medical Journal 298:1617–1618.
Labbok, M.H., and Hendershot, G.E. (1987) Does breast-feeding protect against malocclusion? An analysis of the 1981 Child Health Supplement to the National Health Interview Survey. American Journal of Preventive Medicine 3:227–232.
Lawrence, R.A. (1994) Breastfeeding: A Guide for the Medical Profession, 4th edition. St. Louis, MO: C.V. Mosby Company.
Lee, P.C., Majluf, P., and Gordon, I.J. (1991) Growth, weaning and maternal investment from a comparative perspective. Journal of the Zoological Society of London 225:99–114.
Love, P. (1991) Emotional Incest Syndrome: What to do When a Parent’s Love Rules Your Life. New York: Bantam.
Marmot, M.G., Page, C.M., Atkins, E., and Douglas, J.W.B. (1980) Effect of breastfeeding on plasma cholesterol and weight in young adults. Journal of Epidemiology and Community Health 34:164–167.
Mayer, E.J., Hamman, R.F., Gay, E.C., Lezotte, D.C., Savitz, D.A., and Klingensmith, G.J. (1988) Reduced risk of IDDM among breastfed children. The Colorado IDDM Registry. Diabetes 37:1625–1632.
McKee, L. (1984) Sex differentials in survivorship and the customary treatment of infants and children. Medical Anthropology 8:91–103.
Micozzi, M.S. (1995) Breast cancer, reproductive biology, and breastfeeding. In Breastfeeding: Biocultural Perspectives, ed. P. Stuart-Macadam and K.A. Dettwyler, pp. 347–384. New York: Aldine de Gruyter.
Mitchell E., Scragg, R., Stewart, A.W., Becroft, D.M., Taylor, B.J., Ford, R.P., Hassall, I.B., Barry, D.M., Allen, E.M., and Roberts, A.P. (1991) Results from the first year of the New Zealand Cot Death Study. New Zealand Medical Journal 104:71–76.
Mølbak, K., Gottschau, A., Aaby, P., Højlyng, N., Ingholt, L., and da Silva, A.P.J. (1994) Prolonged breast feeding, diarrhoeal disease, and survival of children in Guinea-Bissau. British Medical Journal 308:1403–1406.
Newcomb, P.A., Storer, B.E., Longnecker, M.P., Mittendorf, R., Greenberg, E.G., Clapp, R.W., Burke, K.P., Willett, W.C., and MacMahon, M. (1994) Lactation and reduced risk of premenopausal breast cancer. New England Journal of Medicine 330:81–87.
Nubé, M., and Asenso-Okyere, W.K. (1996) Large differences in nutritional status between fully weaned and partially breast fed children beyond the age of 12 months. European Journal of Clinical Nutrition 50:171–177.
Osborn, G.R. (1968) Stages in development of coronary disease observed from 1500 young subjects. Relationship of hypotension and infant feeding to aetiology. Colloque international sur le role de la paroi arterielle dan l’atherogenese, pp. 93–129. Paris.
Pabst, H.F., and Spady, D.W. (1990) Effect of breast-feeding on antibody response to conjugate vaccine. Lancet 336:269–270.
Palmer, B. (1998) The influence of breastfeeding on the development of the oral cavity: A commentary. Journal of Human Lactation 14(2):93–98.
Palmer, B. (1999) Breastfeeding: Reducing the risk of obstructive sleep apnea. Breastfeeding Abstracts 18(3):19–20.
Peyton, K.J. (1996) Postpartum Depression in the Absence of Lactation: An Evolutionary Perspective. M.A. Thesis, Department of Anthropology, Texas A&M University, College Station, TX 77843-4352.
Pisacane, A., Buffolano, W., Grillo, G., and Gaudiosi, C. (1992) Infant feeding and Schönlein-Henoch purpura. Acta Paediatrica Scandinavica 81:630.
Pisacane, A., Impagliazzo, N., Russo, M., Valiani, R., Mandarini, A., Florio, C., and Vivo, P. (1994) Breastfeeding and multiple sclerosis. British Medical Journal 308:1411–1412.
Prescott, J. (1990) Affectional bonding for the prevention of violent behaviors: Neurobiological, psychological, and religious/spiritual determinants. In Violent Behavior Volume I: Assessment and Intervention, ed. L.J. Hertzberg, G.F. Ostrum, and J.R. Field, pp. 110–142. New York: PMA Publishing Group.
Prescott, J. (1996) How breast feeding protects newborns and culture: Brain development and affectional bonding for the prevention of depression and violence. Presented at the 1996 Childbirth Congress, ASPO/Lamaze, March 22–25, 1996, Mt. Madonna Center, Watsonville, California.
Rath, F.H., Jr., and Okum, M.E. (1995) Parents and children sleeping together: Cosleeping prevalence and concerns. American Journal of Orthopsychiatry 65:411–418. (See also Dettwyler, 1997).
Rogan, W.J., and Gladen, B.C. (1993) Breast-feeding and cognitive development. Early Human Development 31:181–193.
Romieu, I., Hernandez-Avila, M., Lazcano, E., Lopez, L., and Romero-Jaime, R. (1996) Breast cancer and lactation history in Mexican women. American Journal of Epidemiology 143:543–552.
Ross Labs (1993) Breastfeeding: Trends in Incidence and Duration. Ross Laboratories-sponsored seminar held at the University of Texas School of Nursing on April 21, 1993.
Smith, B.H. (1989) Dental development as a measure of life history in primates. Evolution 43:683–688.
Smith, B.H. (1991a) Age of weaning approximates age of emergence of the first permanent molar in nonhuman primates (abstract). American Journal of Physical Anthropology Supplement 12:163–164.
Smith, B.H. (1991b) Dental development and the evolution of life history in the Hominidae. American Journal of Physical Anthropology 86:157–174.
Smith, B. H. (1992) Life history and the evolution of human maturation. Evolutionary Anthropology 1:134–142.
Sowers, M., Randolph, J., Shapiro, B., and Jannausch, M. (1995) A prospective study of bone density and pregnancy after an extended period of lactation with bone loss. Obstet. Gynecol. 85:285–289.
Stevens, L.J., Zentall, S.S., Deck, J.L., Abate, M.L., Watkins, B.A., Lipp, S.R., and Burgess, J.R. (1995) Essential fatty acid metabolism in boys with attention-deficit hyperactivity disorder. American Journal of Clinical Nutrition, 62:761–768.
Sugarman, M., and Kendall-Tackett, K.A. (1995) Weaning ages in a sample of American women who practice extended breastfeeding. Clinical Pediatrics December:642–647.
Tanoue, Y., and Oda, S. (1989) Weaning time of children with infantile autism. Journal of Autism and Developmental Disorders 19:425–434.
Treloar, S.A., and Martin, N.G. (1990) Age at menarche as a fitness trait: Nonadditive genetic variance detected in a large twin sample. American Journal of Human Genetics 47:137–148.
Truswell, A.S., and Hansen, J.D.L. (1976) Medical research among the !Kung. In Kalahari Hunter-Gatherers: Studies of the !Kung San and Their Neighbors, ed. R.B. Lee and I. DeVore, pp. 166–195. Cambridge, MA: Harvard University Press.
van Lawick-Goodall, J. (1973) The behavior of chimpanzees in their natural habitat. American Journal of Psychiatry 130:1–12.
Virtanen, S.M., Rasanen, L., Aro, A., Lindstrom, J., Sippola, H., Lounamaa, R., Toivanen, L., Tuomilehto, J., and Akerblom, H.K. (1991) Infant feeding in Finnish Children ................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- indiana state university online degrees
- indiana state university majors list
- indiana university spring break 2020
- indiana university schedule 2020
- indiana university bloomington calendar 2020
- indiana university fall 2020 calendar
- indiana university bloomington calendar 2019
- indiana university bloomington academic calendar
- indiana university chemistry ranking
- indiana university ceeb code
- indiana university fafsa deadline 2020
- indiana university fafsa deadline